Although work disability rates associated with osteoarthritis (OA) are high, this problem has received little investigative attention, possibly because it affects older workers, and until recently, premature retirement among older workers was of limited concern. Due to increased longevity, there is now greater interest in preserving work ability among older workers. This project will assess the rates of work disability (work loss and work limitations) and the factors associated with this work disability in the presence of knee OA. A cross-sectional investigation of work disability rates will be carried out in a sample of subjects with and without knee OA: this part of the project has been fully funded through another source. NIH funding is sought in this application to fully carry out the second part of the project, a longitudinal investigation of the impact of knee OA on work and the role of factors associated with the development of work disability. The specific aims relating to the application are: 1) to recruit and characterize the study sample; this sample will be recruited from two ongoing studies of the knee OA, a large HMO and outpatient clinics of the Boston Medical Center; 2) to assess the roles of explanatory variables as factors associated with work limitation in a cross-sectional analysis using baseline data; 3) to describe the rate of work loss over 3 years of follow-up; 4) to describe how work limitation changes over 3 years of follow-up; 5) to assess the roles of possible explanatory variables as predictors of work loss 3 years after baseline assessment; and 6) to assess the roles of possible explanatory variables as predictors of change in work limitation over time. The sample will consist of 344 employed individuals with documented symptomatic knee OA. The outcome variables are work loss and work limitations. Work loss is stopping work for any reason. Work limitation will be measured by the Work Limitations Questionnaire, a survey instrument developed to assess the impact of health on paid work. Explanatory variables are sociodemographic, health, psychosocial, occupational, job, and employer characteristics, out of work conditions, OA medical care and regimens, and job accommodations. Mail survey methods will be used to collect data from all subjects. Primary methods of analysis include multivariate regression, proportional hazards regression, and mixed effects linear regression. This study will provide information about whom to target for knee OA-related work disability interventions and indicate the kinds of interventions that are likely to be needed to reduce work disability in this population.